ClawHub
Back to skill

Security audit

Turn product images and campaign copy into polished seasonal promo posters.

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed poster-generation workflow that uses the user’s Mew API key and product assets for its stated purpose, with some credential/privacy cautions but no artifact-backed malicious behavior.

Install only if you are comfortable giving the agent a mew.design API key and sending product images/campaign copy to Mew’s API. Prefer public URLs you control, avoid approving temporary third-party uploads for confidential assets, and use a key you can rotate or revoke.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (5)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill instructs the agent to write files, execute shell commands, and make network requests, yet declares no permissions. This creates a transparency and policy-enforcement gap: a user or host may approve installation believing it is low-privilege while the skill can handle API keys, write request payloads to disk, and transmit data externally.

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding
The public description says the skill turns product images and copy into posters, but the implementation also validates user API keys and performs outbound authentication-related requests. That mismatch undermines informed consent because users are not clearly told that their key will be tested and their assets transmitted to an external service as part of operation.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The install intent examples are broad enough to match generic requests like 'install this skill' or 'help me install this ClawHub skill,' which can cause unintended installation in ambiguous contexts. In an agent ecosystem, overbroad triggers increase the risk of prompt hijacking or accidental invocation of capabilities involving network access and handling of sensitive API keys.

External Transmission

Medium
Category
Data Exfiltration
Content
Then generate the image:

```bash
curl -sS -X POST "https://api.mew.design/open/api/design/generate" \
  -H "Content-Type: application/json" \
  -H "x-api-key: USER_PROVIDED_KEY" \
  --data @/tmp/holiday-promo-body.json
Confidence
92% confidence
Finding
curl -sS -X POST "https://api.mew.design/open/api/design/generate" \ -H "Content-Type: application/json" \ -H "x-api-key: USER_PROVIDED_KEY" \ --data

External Transmission

Medium
Category
Data Exfiltration
Content
Then generate the image:

```bash
curl -sS -X POST "https://api.mew.design/open/api/design/generate" \
  -H "Content-Type: application/json" \
  -H "x-api-key: USER_PROVIDED_KEY" \
  --data @/tmp/holiday-promo-body.json
Confidence
92% confidence
Finding
https://api.mew.design/

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
SKILL.md:209